Comparison of predictors of hip fracture and mortality after hip fracture in community-dwellers with and without Alzheimer’s disease – exposure-matched cohort study

Anna-Maija Tolppanen, Heidi Taipale, Antti Tanskanen, Jari Tiihonen, Sirpa Hartikainen

Dementia, with Alzheimer's disease (AD) being the most common form, is a major hip fracture risk factor, but currently it is not known whether the same factors predict hip fracture among persons with and without dementia/AD. We compared the predictors of hip fracture and mortality after hip fracture in persons with and without AD. An exposure-matched cohort of all community-dwellers of Finland who received a new clinically verified AD diagnosis in 2005-2011 and had no history of previous hip fracture (N = 67,072) and an age, sex, and region-matched cohort of persons without AD (N = 67,072). Associations between sociodemographic characteristics, comorbidities and medications and risk of hip fracture and mortality after hip fracture were assessed with Cox regression. As expected, the incidence of hip fractures in 2005-2012 (2.19/100 person-years vs 0.90/100 person-years in the non-AD cohort), as well as mortality after hip fracture (29/100 person-years vs 23/100 person-years in the non-AD cohort) were higher in the AD cohort. This difference was evident regardless of the risk factors. Mental and behavioural disorders (adjusted hazard ratio; HR 95% confidence interval CI: 1.16, 1.09-1.24 and 1.71, 1.52-1.92 in the AD and non-AD-cohorts), antipsychotics (1.12, 1.04-1.20 and 1.56, 1.38-1.76 for AD and non-AD-cohorts) and antidepressants (1.06, 1.00-1.12 and 1.34 1.22-1.47 for AD and non-AD-cohorts) were related to higher, and estrogen/combination hormone therapy (0.87, 0.77-0.9 and 0.79, 0.64-0.98 for AD and non-AD-cohorts) to lower hip fracture risk in both cohorts. Stroke (1.42, 1.26-1.62), diabetes (1.13, 0.99-1.28), active cancer treatment (1.67, 1.22-2.30), proton pump inhibitors (1.14, 1.05-1.25), antiepileptics (1.27, 1.11-1.46) and opioids (1.10, 1.01-1.19) were associated with higher hip fracture risk in the non-AD cohort. Similarly, the associations between mortality risk factors (age, sex, several comorbidities and medications) were stronger in the non-AD cohort. AD itself appears to be such a significant risk factor for hip fracture, and mortality after hip fracture, that it overrules or diminishes the effect of other risk factors. Thus, it is important to develop and implement preventive interventions that are suitable and effective in this population.